to unfounded optimism is to insist on realistic assessments based on
external benchmarks. Nobel prize winner Daniel Kahneman calls this
“taking the outside view.” For instance, a management team might
believe and claim a 60 percent chance of success for a new product,
based purely on an internally focused assessment. But of all new-
product launches surveyed each year, only about 10 to 15 percent are
successfully being sold two years later. Even if impressive internal
factors justify elevating the product’s success rate to, say, three times
this base rate, a realistic revised probability is only 30 to 45 percent—a
far cry from 60 percent.518 Chapter 12 Decision Making under UncertaintyThe BP Oil Spill
RevisitedOn April 20, 2010 in the Gulf of Mexico, the Deepwater Horizondrilling rig
exploded, killing 11 workers and unleashing one of the largest oil leaks and
environmental disasters in U.S. history. BP had leased the rig to drill an
exploratory well at a water depth of approximately 5,000 feet and was in the
process of temporarily shutting down the well. Offshore drilling at great depth
involves a myriad of operational risks. According to the preliminary report of
the commission investigating the disaster, BP and its contractors Halliburton
Co. and Transocean Ltd. repeatedly took risky actions in the interests of saving
time and money. Leading up to and precipitating the disaster, the three par-
ties—due to poor communications, conflicting assessments, and confusion
about responsibilities—made a series of costly mistakes and misjudgments. The
available evidence indicates that the explosion was not an unavoidable fluke but
rather a result of management decisions stemming from a culture within BP
that downplayed safety risks.^4 Only on September 9, after repeated failed
attempts to stem the leak, was the well finally shut down.
The overview of the BP oil spill in Chapter 1 concluded with a key question:
What decision-making pitfalls and mistakes contributed to the accident?
Research in behavioral economics points to a number of psychological decision
traps as potential contributors to BP’s excessive risk taking. In highly uncer-
tain settings, self-serving overoptimismand overconfidencecan blind man-
agement to the true risks they face. Given a 20-year record of aggressive oil
exploration with many successes and few mishaps, BP’s top management could
easily convince itself that its margins of safety (even if not so high by industry
standards) were more than adequate. Overoptimism would lead BP to see a
spill risk as a less than “one in a thousand” probability, whereas the objective
risk might be closer to one in one hundred. Overconfidence could convince(^4) This account is based on numerous published sources, including “Report on Causes of the
Deepwater Horizon Oil Rig Blowout and Ways to Prevent such Events,” U.S. Department of the
Interior, November 16, 2010; G. Chazan, “BP’s Safety Drive Faces Rough Road,” The Wall Street
Journal(February 1, 2011), p. A1; S. Lyall, “In BP’s Record, a History of Boldness and Costly
Blunders,” The Wall Street Journal(July 13, 2010), p. A1; J. Nocera, “BP Ignored the Omens of
Disaster,” The New York Times(June 19, 2010), p. B1; and I. Urbina, “In Gulf, It was Unclear Who
Was in Charge of Rig,” The New York Times(June 6, 2010), p. A1.
c12DecisionMakingunderUncertainty.qxd 9/29/11 1:34 PM Page 518